SC²S Colloquium - April 30, 2012
|Date:||April 30, 2012|
|Time:||3:00 pm, s.t.|
Vladimir Golkov: Kurtosis Estimation in Diffusion Spectrum Imaging Using Non-Gaussian Noise Models
Diffusion magnetic resonance imaging (diffusion MRI) is a non-invasive imaging method that allows estimating the molecular self-diffusion of water molecules within the surrounding biological tissue, and determining the macroscopic orientation of the underlying microscopic cellular architecture. Due to the shortcomings of the popular diffusion model called diffusion tensor imaging (DTI), various alternative diffusion models and corresponding acquisition schemes have been proposed that reflect the histological tissue architecture more closely. Since diffusion MRI has yielded promising experimental results over the last years, our aim is to contribute to diffusion MRI methods that may improve the diagnosis of multiple sclerosis, traumatic brain injury, and other conditions. In the present work, we have defined four quality criteria that a diffusion MRI protocol should fulfill. Each step of the protocol, from data acquisition all the way to calculation of derived measures, influences the results. Focusing on model fitting and on calculation of derived measures, we have used the defined quality criteria and a simulated data model to compare the quality of several variants of biased and unbiased fitting methods based on diffusion spectrum imaging (DSI) data acquisition scheme, while using diffusion kurtosis imaging (DKI) as an advanced diffusion model. As a result, this comparison demonstrates how our quality criteria can help to point out optimal data processing approaches.